Draft genome sequences of four Staphylococcus hyicus strains, SC302, SC304, SC306, and SC310, isolated from swine from Eastern Canada

ABSTRACT The bacterium Staphylococcus hyicus causes porcine exudative epidermitis in piglets, which represents both health and welfare concerns. Few genome sequences of this pathogen are published. We provide four additional ones to help future genomic analysis of S. hyicus. These are genomes of strains isolated from Canadian swine.

S taphylococcus hyicus is the causative agent of porcine exudative epidermitis (EE)   in piglets, which affects swine farms sporadically, as S. hyicus is endemic in most herds worldwide (1).Despite the problems caused, few S. hyicus genome sequences are currently available.We aim to better understand S. hyicus by increasing available genome sequences.
We present draft genome sequences of four S. hyicus strains isolated from East ern Canada.The strains (SC302, SC304, SC306, and SC310) were isolated and initially identified by the Ministère de l'Agriculture, des Pêcheries et de l'Alimentation du Québec (MAPAQ) veterinary diagnostic services in 2021, following necropsy of piglets submitted for clinical signs of EE and/or lameness.Tissue samples (Table 1) were plated onto 5% sheep blood agar, incubated at 35°C in 5% CO 2 .Colonies selected based on morphology were identified using MALDI-TOF mass spectrometry (MALDI biotyper smart, Bruker Daltonics, Bremen, Germany).Confirmed S. hyicus colonies were streaked onto blood agar to ensure purity before being inoculated into tryptic soy broth (TSB) + 20% glycerol and frozen at −80°C.
After recovery from frozen stocks, plating on TSA, and incubation at 37°C for 24 hours, an inoculum of each strain was grown in TSB at 37°C, 200 rpm, for subsequent genomic DNA extraction using DNeasy Blood and Tissue Kits (Qiagen, Canada), according to the manufacturer's instructions.The purified DNA was prepared into sequencing libraries using NEBNext Ultra II FS DNA Library Prep Kits (Ipswich, MA, USA) and sequenced using a MiSeq instrument system (Illumina, San Diego, CA, USA) generating 2 × 300 bp reads (Plateforme d'Analyses Génomiques, IBIS, Université Laval).Sequencing reads were verified with FastQC version 0.11.8 (2), filtered using fastp version 0.23.2 (3), and de novo assembled in contigs using Shovill version 1.1.0(4).This Whole-Genome Shotgun project has been deposited at DDBJ/EMBL/GenBank under BioProject PRJNA985240.Default parameters were used for all bioinformatics tools, unless otherwise specified.Assembly statistics are presented in Table 1.
Taxonomic identification was performed by molecular phylogeny (Fig. 1) including the 22 S. hyicus genome sequences currently available on GenBank, 1 reference strain for each of 64 other Staphylococci species, and 3 closely related genera as outgroup.
The phylogeny was coupled for S. hyicus strains with average nucleotide identity (ANIm) analysis and calculation of a matrix of the percentage of conserved proteins (POCP) values.POCP results are especially useful to show distinctiveness of strains.
The four genomes presented are from Eastern Canada, where swine industry is important.Interestingly, POCP results suggest diversity in these strains despite being from the same region and with other Canadian strains isolated from cattle (strains SNUC XXXX, Fig. 1).Strains CCUG 6509, NCTC 10350, and ATCC 11249 regroup as they come from the same original isolate (10)(11)(12).Another group forms between isolates from same BioProject (strains XXXRC1), suggesting clonality or close link.
Sequencing these strains was partly aimed to analyze their possible resistance to tetracycline, which is largely used in swine industry (13).We found two distinct

FIG 1
FIG 1 Molecular phylogeny of the genus Staphylococcus with emphasis on species hyicus.Orthologous genes were identified with COG and OMCL through GET_HOMOLOGUES version 20220822 (5).The sequences of the 1,003 orthologous gene clusters were aligned with mafft version 7.515 (6) and TranslatorX and finally filtered with BMGE version 1.12 (7).The phylogenetic tree was then made with IQ-TREE version 2.2.0 (8), with 10,000 ultrafast bootstraps, choosing the best model for each of the 1,003 partitions.The four strains of S. hyicus whose genomes were sequenced by the present study are in red.The ANIm matrix allows evaluating the nucleotide diversity, and the POCP values are based on the orthologous proteins shared between the strains.The values of ANIm and POCP were computed using pyani version 0.2.12 (9) and GET_HOMOLOGUES version 22082022 (5), respectively.Bootstrap values are shown when less than 100.Strain NCTC 8294 (BioSample number SAMEA3499053) was revealed to be a S. aureus strain (red asterisk) in the phylogeny which explains its exclusion from ANIm and POCP analysis.